Using a sliding window over per-token Shannon entropy from GPT-4o log-probabilities, the paper flags high-uncertainty spans that, in a 36-image proof-of-concept, tend to contain most math OCR errors.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Entropy Heat-Mapping: Localizing GPT-Based OCR Errors with Sliding-Window Shannon Analysis
Using a sliding window over per-token Shannon entropy from GPT-4o log-probabilities, the paper flags high-uncertainty spans that, in a 36-image proof-of-concept, tend to contain most math OCR errors.